What happens when we lose global health data?

Vox

2 days ago

is a fellow for Future Perfect. He reports on global health, science, and biomedicine, focusing on how policies and systems shape progress.
A census enumerator, right, talks with a Maasai woman during the population and housing census, the first time being conducted digitally, at a village in Engikaret on August 23, 2022. AFP via Getty Images
When President Donald Trump and Elon Musk fed the US Agency for International Development into the wood chipper earlier this year, one of the lesser-known casualties was the shutdown of an obscure but crucial program that tracked public health information on about half of the world's nations.
For nearly 40 years, the Demographic and Health Surveys (DHS) Program has served as the world's health report card. In that time, it has carried out over 400 nationally representative surveys in more than 90 countries, capturing a wide range of vital signs such as maternal and child health, nutrition, education levels, access to water and sanitation, and the prevalence of diseases like HIV and malaria.
Taken together, it offered perhaps the clearest picture ever compiled of global health.
And that clarity came from how rigorous these surveys were. Each one started with a globally vetted blueprint of questions, used by hundreds of trained local surveyors who went door-to-door, conducting face-to-face interviews in people's homes. The final, anonymized data was then processed by a single contractor ICF International, a private consulting firm based in Reston, Virginia, which made the results standardized and comparable across countries and over time. Its data powered global estimates from institutions like the Institute for Health Metrics and Evaluation, which in turn shaped public health policy, research, and funding decisions around the world. 'If DHS didn't exist, comparing anemia across countries would be a PhD thesis,' said Doug Johnson, a senior statistician at the nonprofit IDinsight.
Crucially, DHS also tracked things few other systems touched, like gender-based violence, women's autonomy, and attitudes toward domestic abuse. Doctor's offices aren't representative and only capture folks who can access a formal health care system. Also, since DHS data is anonymized, unlike a police report, responders don't have to fear intervention if they don't want it. 'You can't get answers from other sources to sensitive questions like the ones DHS posed,' said Haoyi Chen from the UN Statistics Division, pointing to one example: Is a husband justified in beating his wife if she burns the food?
Then, earlier this year, DHS was shut down.
The decision came as part of the Rescissions Act of 2025, a bill passed in June that clawed back $9.4 billion from foreign aid and other programs. Eliminating DHS saved the government some $47 million a year — only about 0.1 percent of the total US aid budget, or half the cost of a single F-35 fighter jet.
Future Perfect
Explore the big, complicated problems the world faces and the most efficient ways to solve them. Sent twice a week. Email (required)
Sign Up
By submitting your email, you agree to our Terms and Privacy Notice . This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
That tiny budget cut has had immediate consequences. The move halted around 24 in-progress country surveys – 10 of which were just short of final publication, and three in Ethiopia, Guinea, and Uganda that were stopped mid-fieldwork. The program's public-facing website remains up, but the machinery behind it is gone. With no one to approve new applications, the process for researchers to access the underlying microdata has ground to a halt.
How the DHS has saved lives
The shutdown isn't just about numbers on a spreadsheet. Here's how DHS data has shaped policy and saved lives across the globe. Guinea: DHS data was used to help tailor the rollout of the new malaria vaccine
India: The 2019–2021 national survey (India's version of the DHS) showed a stark gap in menstrual hygiene between urban and rural areas, which prompted a new national policy to address the disparity.
Nepal: A 2016 DHS survey revealed stagnating maternal mortality rates. This spurred the government to enhance its Safe Motherhood Program , resulting in more women delivering babies in health facilities rather than homes — and fewer women dying in childbirth.
Nigeria: DHS surveys showed child marriage rates as high as 76 percent in some states. Advocates used that as evidence to successfully push local governments to strengthen their laws against the practice.
There will also be long-term damage. When governments or aid organizations can no longer see exactly where children are malnourished, where malaria outbreaks are quietly spreading, or where mothers are dying in childbirth, they can't effectively target life-saving interventions, leaving the most vulnerable populations to pay the price. For 24 countries, including the Democratic Republic of Congo and Mali, the DHS was the sole data source for the UN's official maternal mortality estimates. Going forward, 'it would just be basically estimates that are based on other countries' data,' says Saloni Dattani, a editor on science and global health at Works in Progress magazine and 2022 Future Perfect 50 honoree. 'We just wouldn't know.'
Without the data DHS provided, foreign aid becomes less effective, and less accountable 'We have no way of externally or objectively estimating the positive impact that those [aid] programs are having, or negative,' said Livia Montana, the former deputy director of the DHS Program, who is now a survey director for the Understanding America Study at the University of Southern California.
Naturally, the global health community has been scrambling to plug the enormous gap. The Gates Foundation recently committed $25 million in emergency funding to rescue some ongoing surveys, and Bloomberg Philanthropies has also stepped in with a separate commitment to support the effort.
This funding is a crucial lifeline, but only a stopgap. The search for a long-term fix has forced a reckoning with the old programs' flaws. Everyone agrees that DHS delivered high-quality, trusted data — but it wasn't perfect. Many experts have criticized it as fundamentally 'donor-driven,' with priorities that didn't always align with the national interests of the countries it surveyed. For instance, the program's historic focus on reproductive health was a direct reflection of the priorities of its primary funder, USAID, and some country officials privately felt the data served the accountability needs of international organizations better than their own immediate planning needs.
This has created a central dilemma for the global development community: is it possible to build a new system that is both genuinely country-led and also globally comparable?
A lifeline and a reckoning
Faced with this data vacuum, an obvious question arises: Why can't other global organizations like the World Health Organization or the United Nations simply step in and take over?
It's not out of the question, but it would be really, really difficult. Think of it this way: The DHS Program was like a single, powerful architecture firm that perfected a blueprint and built houses in 90 neighborhoods for 40 years. Because it was a single program managed by private contractor, ICF International, and backed by one major funder, USAID, it could enforce a standardized methodology everywhere it worked. As a for-profit firm, ICF's interest was also financial, it managed the global contract and profited from the work.
The UN and WHO, by contrast, act as the global city planners: Their mandate isn't to design and build the houses themselves, but to set the building codes and safety standards for everyone. According to WHO, its role is not to 'directly fund population-based surveys,' but to provide leadership and bring the right stakeholders together.
While that mandate may prevent the UN from simply inheriting the old program's work, it makes it an ideal coordinator for the path forward, says Caren Grown, a senior fellow at the Brookings Institution's Center for Sustainable Development. Grown argues that the UN is the only body that can handle the 'heavy lift' of coordinating all the different countries, donors, and organizations.
And now that the DHS has been dissolved, both Grown and Chen are now part of a UN task force attempting to establish new internationally agreed-upon standards for how health data should be collected and governed.
At the same time, other efforts are more focused on the practical work of implementation rather than on global governance. Montana is leading a coalition to 'rebuild elements of DHS' by creating a global consortium of research institutions that can provide technical support to countries. These efforts were catalyzed by initial conversations hosted by organizations like the Population Reference Bureau, which brought together donors, government agencies, and global data users to grapple with the shutdown's immediate aftermath.
Critics argue that for every India, there are a dozen other nations where the program's sudden collapse is proof that a deep, sustainable capacity was never built.
Between this mishmash, the most practical development has been a lifeline from the Gates Foundation, which announced a $25 million investment in 'bridge funding.' Separately, in a statement to Vox, Bloomberg Philanthropies confirmed its commitment to fund the completion of an additional 12-country surveys over the next eight months. A source from the Gates Foundation clarified that Bloomberg's commitment is on top of theirs, confirming the two are distinct but coordinated rescue efforts.
The Gates Foundation framed its effort as a temporary, stabilizing measure designed to give the global health community a much-needed respite. 'We believe data is — and must remain — a global public good,' said Janet Zhou, a director focused on data and gender equality at the Gates Foundation. 'Our interim support is helping to stabilize 14 ongoing country surveys. … This investment is designed to give global partners and national governments the time and space needed to build a more sustainable, country-led model for health data.'
That support is aimed at the most urgent work: finishing surveys that were nearly complete, like in Ethiopia, and reopening the four-decade-old data archive. But rather than giving each respective country the money to complete their ongoing surveys, the Gates funding will be administered by ICF International, the same for-profit firm that ran the original DHS.
The decision to work with the existing contractor, ICF International, was a pragmatic one. Continuing with the same implementer was the 'quickest, most affordable way' to prevent waste, and 'multiple host countries have shared a preference' to complete their work with the firm, said a source at the Gates Foundation.
A Sudanese mother sits with her children at a shelter in the al-Qanaa village in Sudan's southern White Nile state on September 14, 2021. Ashraf Shazly/AFP via Getty Images
It's a powerful argument for triage in an emergency, but it also papers over deeper flaws. Take a look at Nigeria, for example: Fieldwork for its 2023–'24 DHS finished in May 2024, and the questionnaires gathered new estimates of maternal and child deaths. Nigeria also ran a separate study to probe exactly why mothers and children are dying. In principle, the two datasets should dovetail but beyond a headline-numbers report, the full DHS micro-dataset is still in ICF's processing queue — likely frozen after DHS's shuttering.
That bottleneck illustrates what critics mean by 'donor-driven.' With barely 3 percent of household surveys in low-income countries fully-financed by the local government, the WHO notes, most nations must rely on 'externally led surveys…limiting continuity and national ownership.' When the donor funding stops, so does the data pipeline.
An ICF spokesperson pushed back saying survey priorities were 'primarily shaped by the participating countries.' Yet, of the $25 million that arrived from Gates, a large portion of it will go toward completing large-scale surveys in Nigeria and Kenya, two countries that also happen to be key 'geographies of interest' for the Gates Foundation's own strategic priorities, underscoring how funders still steer the spotlight.
Insiders I spoke with described ICF's system as a 'black box,' with key parts of its methodology controlled by the contractor, leaving countries without the capacity to stand on their own. That matters because without home-grown statisticians and know-how, ministries can't rerun surveys or update indicators without outside help. In response, ICF stated that the program has a 'proven track record of building a long-term capacity,' noting that countries like India no longer require its assistance.
But critics argue that for every India, there are a dozen other nations where the program's sudden collapse is proof that a deep, sustainable capacity was never built. This dependency creates a fragile system that can, as just happened, collapse overnight, leaving countries unable to continue that work on their own.
This unresolved tension brings the debate back to a central question from the UN's Chen. 'DHS has been there for four decades,' she asks, 'and why are we still having this program doing the survey for countries?'
Chen's question gets to the heart of the debate. But grappling with the flaws of the past can't get in the way of surviving the present. Existing global health data is already several years out of date due to the pandemic, while crises in maternal mortality and child nutrition continue to unfold. The need is for reliable data now, because the fundamental reality remains: You can't help people you can't see.